Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!rutgers!mit-eddie!uw-beaver!milton!ogicse!mintaka!spdcc!tauxersvilli!alphalpha!nazgul From: nazgul@alphalpha.com (Kee Hinckley) Newsgroups: comp.sys.apollo Subject: Re: Another NFS gotcha Message-ID: <1991Mar23.053057.14306@alphalpha.com> Date: 23 Mar 91 05:30:57 GMT References: <508478b3.1bc5b@pisa.citi.umich.edu> Organization: asi Lines: 25 In article <508478b3.1bc5b@pisa.citi.umich.edu> rees@citi.umich.edu (Jim Rees) writes: >In article , hanche@imf.unit.no (Harald Hanche-Olsen) writes: > > % tar cvf /dev/null fixes > tar: ERROR!! Cannot archive fixes. It is of type nfs_dir. > >This isn't a bug in nfs, it's a bug in tar. Apollo tar is badly broken and Another thing to watch out for. Tar isn't too hot on closing files. We have a distributed software development system that works over UUCP and uses tar to transfer updates. Often we'll ship 50 or 60 source files along with an equal number of control files. But if we exceed 100 or so files tar dies - having run out of streams. You might try /usr/apollo/bin/pax. It can do both tar and cpio archives. -kee -- Alfalfa Software, Inc. | Poste: The EMail for Unix nazgul@alfalfa.com | Send Anything... Anywhere 617/646-7703 (voice/fax) | info@alfalfa.com I'm not sure which upsets me more: that people are so unwilling to accept responsibility for their own actions, or that they are so eager to regulate everyone else's.